BED-9721 migrate GitHub entity panels into schema

pyproject.toml

@@ -28,7 +28,7 @@ local_scheme = "no-local-version"
 
 [dependency-groups]
 dev = [
-    "openhound>=0.2.7",
+    "openhound>=0.4.0",
     "pre-commit>=4.5.1",
     "pytest>=9.0.1",
     "ruff>=0.15.5",

src/openhound_github/graph.py

@@ -4,6 +4,9 @@
 from openhound.core.models.entries_dataclass import Node as BaseNode
 from openhound.core.models.entries_dataclass import NodeProperties as BaseProperties
 
+GITHUB_SOURCE_KIND = "GitHub"
+SAML_SOURCE_KIND = "SAML"
+
 
 @dataclass
 class GHNodeProperties(BaseProperties):
@@ -22,6 +25,13 @@ def __post_init__(self):
         # Use GitHub's native node_id as the OpenGraph node id so edges can
         # reference nodes by the same identifier used during collection.
         self.id = self.properties.node_id
+        source_kind = (
+            SAML_SOURCE_KIND
+            if any(kind.startswith("SAML_") for kind in self.kinds)
+            else GITHUB_SOURCE_KIND
+        )
+        if source_kind not in self.kinds:
+            self.kinds.append(source_kind)
 
 
 @dataclass

src/openhound_github/main.py

@@ -8,7 +8,7 @@
 from .lookup import GithubLookup
 from .transforms import transforms
 
-app = OpenHound("github", source_kind="GitHub", help="OpenGraph collector for GitHub")
+app = OpenHound("github", help="OpenGraph collector for GitHub")
 
 
 @app.collect()

src/openhound_github/models/scim_user.py

@@ -23,6 +23,8 @@
 from openhound_github.main import app
 from openhound_github.models.saml_helpers import detect_foreign_idp
 
+SCIM_SOURCE_KIND = "SCIM"
+
 
 def scim_organization_id(scope_node_id: str) -> str:
     return f"SCIM_Organization_{scope_node_id}"
@@ -40,15 +42,15 @@ class ScimNodeProperties(NodeProperties):
     profile_url: str | None = None
     enterprise: str | None = None
     organization: str | None = None
-    source_kind: str | None = None
 
 
 @dataclass
 class ScimNode(Node):
     id: str
 
     def __post_init__(self):
-        return None
+        if SCIM_SOURCE_KIND not in self.kinds:
+            self.kinds.append(SCIM_SOURCE_KIND)
 
 
 class Name(BaseModel):
@@ -128,7 +130,6 @@ def as_node(self) -> ScimNode:
                 environmentid=self.scope_node_id,
                 enterprise=self.enterprise_slug,
                 organization=self.org_login,
-                source_kind="GitHub",
             ),
         )
 
@@ -210,7 +211,6 @@ def as_node(self) -> ScimNode:
                 profile_url=self.meta.location if self.meta else None,
                 enterprise=self.enterprise_slug,
                 organization=self.org_login,
-                source_kind="GitHub",
             ),
         )
 
@@ -306,7 +306,6 @@ def as_node(self) -> ScimNode:
                 external_id=self.external_id,
                 enterprise=self.enterprise_slug,
                 organization=self.org_login,
-                source_kind="GitHub",
             ),
         )
 

tests/test_runner_models.py

@@ -197,7 +197,7 @@ def test_org_runner_group_keeps_generic_runner_group_label() -> None:
 
     node = group.as_node
 
-    assert node.kinds == [nk.ORG_RUNNER_GROUP, nk.RUNNER_GROUP]
+    assert node.kinds == [nk.ORG_RUNNER_GROUP, nk.RUNNER_GROUP, "GitHub"]
     assert node.properties.scope == "organization"
     assert node.id == "ORG_1_runner_group_1"
 
@@ -267,22 +267,26 @@ def test_runner_groups_and_runners_use_scope_owner_prefixes_with_generic_suffixe
     )
     repo_runner._lookup = SimpleNamespace(org_id_for_login=lambda _login: "ORG_1")
 
-    assert org_runner.as_node.kinds == [nk.ORG_RUNNER, nk.RUNNER]
+    assert org_runner.as_node.kinds == [nk.ORG_RUNNER, nk.RUNNER, "GitHub"]
     assert org_runner.as_node.properties.scope == "organization"
     assert org_runner.as_node.id == "ORG_1_runner_8"
     assert org_runner.as_node.properties.name == "acme/org-runner-1"
     assert org_runner.as_node.properties.displayname == "org-runner-1"
-    assert group.as_node.kinds == [nk.ENTERPRISE_RUNNER_GROUP, nk.RUNNER_GROUP]
+    assert group.as_node.kinds == [
+        nk.ENTERPRISE_RUNNER_GROUP,
+        nk.RUNNER_GROUP,
+        "GitHub",
+    ]
     assert group.as_node.properties.scope == "enterprise"
     assert group.as_node.id == "ENT_1_runner_group_2"
 
-    assert runner.as_node.kinds == [nk.ENTERPRISE_RUNNER, nk.RUNNER]
+    assert runner.as_node.kinds == [nk.ENTERPRISE_RUNNER, nk.RUNNER, "GitHub"]
     assert runner.as_node.properties.scope == "enterprise"
     assert runner.as_node.id == "ENT_1_runner_9"
     assert runner.as_node.properties.name == "acme-enterprise/enterprise-runner-1"
     assert runner.as_node.properties.displayname == "enterprise-runner-1"
 
-    assert repo_runner.as_node.kinds == [nk.REPO_RUNNER, nk.RUNNER]
+    assert repo_runner.as_node.kinds == [nk.REPO_RUNNER, nk.RUNNER, "GitHub"]
     assert repo_runner.as_node.properties.scope == "repository"
     assert repo_runner.as_node.id == "REPO_1_runner_10"
     assert repo_runner.as_node.properties.name == "acme/repo/repo-runner-1"

tests/test_saml_models.py

@@ -1,6 +1,7 @@
 from types import SimpleNamespace
 
 from openhound_github.kinds import edges as ek
+from openhound_github.kinds import nodes as nk
 from openhound_github.graphql import ENTERPRISE_SAML_QUERY, SAML_IDENTITIES_QUERY
 from openhound_github.models.external_identity import ExternalIdentity
 from openhound_github.models.saml_assertion_consumer_service import (
@@ -58,6 +59,7 @@ def _saml_account_edge(identity: ExternalIdentity):
 def test_external_identity_prefers_saml_username_for_display_name() -> None:
     identity = _identity_with_lookup()
 
+    assert identity.as_node.kinds == [nk.EXTERNAL_IDENTITY, "GitHub"]
     assert identity.as_node.properties.name == "Alice@example.com"
     assert identity.as_node.properties.displayname == "Alice@example.com"
 
@@ -106,6 +108,9 @@ def test_normalized_saml_nodes_expose_contract_metadata() -> None:
     assert service_provider.as_node.id == "github:saml:sp:org:acme"
     assert issuer.as_node.id == "github:saml:trusted-issuer:org:acme"
     assert acs.as_node.id == "github:saml:acs:org:acme"
+    assert service_provider.as_node.kinds == [nk.SAML_SERVICE_PROVIDER, "SAML"]
+    assert issuer.as_node.kinds == [nk.SAML_ISSUER, "SAML"]
+    assert acs.as_node.kinds == [nk.SAML_ASSERTION_CONSUMER_SERVICE, "SAML"]
     assert sp_properties.github_deployment_id == DEFAULT_GITHUB_DEPLOYMENT_ID
     assert sp_properties.github_web_origin == DEFAULT_GITHUB_WEB_ORIGIN
     assert sp_properties.schema_contract_version == SAML_CONTRACT_VERSION
@@ -228,6 +233,7 @@ def test_saml_provider_replays_snake_case_fields_and_deployment_metadata() -> No
     assert provider.sso_url == "https://issuer.example.com/sso"
     assert provider.signature_method == "rsa-sha256"
     assert provider.idp_certificate == "certificate-data"
+    assert provider.as_node.kinds == [nk.SAML_IDENTITY_PROVIDER, "GitHub"]
     assert provider.as_node.properties.github_deployment_id == "github.example.com"
     assert provider.as_node.properties.github_web_origin == "https://github.example.com"
 

tests/test_scim_models.py

@@ -90,7 +90,7 @@ def test_scim_user_emits_normalized_edges_without_legacy_correlation_by_default(
     edges = list(user.edges)
     unmatched_edges = list(unmatched_user.edges)
 
-    assert node.kinds == [nk.SCIM_USER]
+    assert node.kinds == [nk.SCIM_USER, "SCIM"]
     assert node.properties.environmentid == "ENT_NODE_1"
     assert node.properties.external_id == "00u-okta-1"
     assert node.properties.name == "alice@example.test"
@@ -136,6 +136,7 @@ def test_scim_group_emits_membership_and_tenant_scoped_legacy_correlation() -> N
 
     edges = list(group.edges)
 
+    assert group.as_node.kinds == [nk.SCIM_GROUP, "SCIM"]
     assert [edge.kind for edge in edges] == [
         ek.SCIM_CONTAINS,
         ek.SCIM_MEMBER_OF,
@@ -180,7 +181,7 @@ def test_scim_organization_stays_within_github_environment_root() -> None:
     node = organization.as_node
 
     assert node.id == "SCIM_Organization_ENT_NODE_1"
-    assert node.kinds == [nk.SCIM_ORGANIZATION]
+    assert node.kinds == [nk.SCIM_ORGANIZATION, "SCIM"]
     assert node.properties.environmentid == "ENT_NODE_1"